209 - Hemodynamically significant patent ductus arteriosus exposure is associated with impaired glucose tolerance among extremely preterm infants in the first 14 postnatal days

Poster Number: 209
Publication Number: 209.220

Melanie C. Morris, University of Iowa, Iowa City, IA, United States; susan j. Carlson, University of Iowa Stead Family Children's Hospital, Iowa City, IA, United States; Patrick J. McNamara, University of Iowa Roy J. and Lucille A. Carver College of Medicine, Iowa City, IA, United States; Regan E. Giesinger, University of Iowa, Iowa City, IA, United States

Background:

Neonatal hyperglycemia is associated with risk of severe intracranial hemorrhage, retinopathy of prematurity, necrotizing enterocolitis, and late onset bacterial infection. Anticipating glucose intolerance may facilitate a reduced hyperglycemia burden. Hemodynamically significant patent ductus arteriosus (hsPDA) alters blood flow patterns and may alter glucose control.

Objective:

To evaluate the relationship between hsPDA and glucose tolerance among extremely preterm infants in the first 2 postnatal weeks.

Design/Methods:

A retrospective analysis of a prospectively collected cohort was conducted of neonates < 27 weeks gestational age with serial targeted neonatal echocardiography (TnECHO) during the first 14 postnatal days. Neonates were classified by hsPDA status based on Iowa PDA score of >6 and receipt of treatment as severe (score ≥10 & ≥7 days), moderate (score 6-10 and/or < 7 days) or no shunt. Metabolic disorders & insulin exposure were exclusions. Daily highest glucose, lowest glucose infusion rate (GIR) and glucose: GIR ratio were calculated. Primary outcome was a difference in average glucose:GIR ratio [peak glucose÷lowest GIR/24h; average over study period] between shunt groups. Change in glucose tolerance over time was compared between a nested cohort who developed hsPDA during the study period compared to those with no shunt. Univariate analysis and generalized estimating equations were used.

Results:

A cohort of 154 patients were screened of whom 141 met eligibility criteria. A median of 4 [3, 5] TnECHOs per patient were performed in the first 14 days. Demographic factors were similar between hsPDA groups aside from a trend towards lesser antenatal steroids among shunt patients [Table 1]. Severe shunt exposure was associated with higher median glucose: GIR ratio and overall burden of hyperglycemia despite no difference in GIR than both moderate shunt and no shunt exposure. A subgroup of 24 patients developed hsPDA during the study period [Table 2]. These patients demonstrated a decline in glucose tolerance as evidenced by increased glucose: GIR ratio (p < 0.001), lesser tolerance of increased GIR (p= < 0.001) and a trend towards higher peak glucose (p=0.063).[Figure 1] In contrast, patients who did not develop shunt displayed improved glucose tolerance over the study period.

Conclusion(s):

hsPDA is associated with impaired glucose tolerance both overall and within individual patients. Hyperglycemia may therefore be an unappreciated manifestation of hsPDA; monitoring for changes in glycemic control during therapy requires further prospective evaluation.

Table 1: Clinical and demographic characteristics of patients exposed vs unexposed to severe (n=42) vs moderate (n=27) hsPDA vs no shunt (n=69) during the first 14 postnatal days.
Table 2: Subgroup analysis of 24 patients who developed new onset hemodynamically significant PDA during the study period.